MediumNLPnlp-087

CLS Classification Head Logits

Problem

Return the logits W x + b for a classification head, where cls_vector is x, weights is one row per class, and bias is one entry per class.

Iterate classes in row order.

Hint

Each class row is a dot product with the pooled vector plus its bias.

case 1

in: [[1,2],[[1,0],[0,1]],[0,0]]

expected: [1,2]

case 2

in: [[1,1],[[1,1]],[0.5]]

expected: [2.5]

case 3

in: [[0,0],[[1,2],[-1,-2]],[1,-1]]

expected: [1,-1]

case 4

in: [[],[[1]],[0]]

expected: [0]

Starter codePython

def cls_classification_logits(cls_vector, weights, bias):
    # Your code here
    pass

4 test cases · run your code against them in the editor